diff options
author | Petr Baudis <pasky@suse.cz> | 2009-07-16 10:10:10 -0700 |
---|---|---|
committer | Ulrich Drepper <drepper@redhat.com> | 2009-07-16 10:10:10 -0700 |
commit | 137028b4d7e50f71906c1656c27079eac5a1d085 (patch) | |
tree | 61b121088c68e50ff2483cd15675729cf3c337e7 /nscd/pwdcache.c | |
parent | 50158f95525ca59459a90f2a7bc65ceb892a0807 (diff) | |
download | glibc-137028b4d7e50f71906c1656c27079eac5a1d085.tar glibc-137028b4d7e50f71906c1656c27079eac5a1d085.tar.gz glibc-137028b4d7e50f71906c1656c27079eac5a1d085.tar.bz2 glibc-137028b4d7e50f71906c1656c27079eac5a1d085.zip |
Fix lock handling in memory hander of nscd.
The commit 20e498bd removes the pthread_mutex_rdlock() calls, but not the
corresponding pthread_mutex_unlock() calls. Also, the database lock is never
unlocked in one branch of the mempool_alloc() if.
I think unreproducible random assert(dh->usable) crashes in prune_cache() were
caused by this. But an easy way to make nscd threads hang with the broken
locking was.
Diffstat (limited to 'nscd/pwdcache.c')
-rw-r--r-- | nscd/pwdcache.c | 6 |
1 files changed, 1 insertions, 5 deletions
diff --git a/nscd/pwdcache.c b/nscd/pwdcache.c index 2338e7e1e0..fc5b44eef0 100644 --- a/nscd/pwdcache.c +++ b/nscd/pwdcache.c @@ -153,8 +153,6 @@ cache_addpw (struct database_dyn *db, int fd, request_header *req, (void) cache_add (req->type, key_copy, req->key_len, &dataset->head, true, db, owner, he == NULL); - pthread_rwlock_unlock (&db->lock); - /* Mark the old entry as obsolete. */ if (dh != NULL) dh->usable = false; @@ -362,12 +360,10 @@ cache_addpw (struct database_dyn *db, int fd, request_header *req, (void) cache_add (GETPWBYUID, cp, key_offset, &dataset->head, false, db, owner, false); } - - out: - pthread_rwlock_unlock (&db->lock); } } +out: if (__builtin_expect (written != total, 0) && debug_level > 0) { char buf[256]; |